Network engineers spend a lot of time translating.
A stakeholder asks a simple question:
"Can this application reach that subnet?"
The answer often requires multiple tools, several queries, and plenty of expertise. Then the follow-up questions arrive, and the process starts over.
With Forward Enterprise 26.5, we're introducing for our SaaS users Forward AI— an intelligent, conversational, and agentic system built natively into the Forward Enterprise platform that lets engineers interact with their network in plain language and get answers grounded in Forward's network digital twin.
For the first time, the network can answer back.
A Skill Multiplier for Network and Security Teams
Networks have become dramatically more complex, but most teams haven't grown at the same pace.
- Forward AI helps engineers do more with the expertise they already have.
- It doesn't replace network knowledge. It removes the friction between a question and an answer.
- Experienced engineers can move faster without jumping between tools. Forward AI also helps make expert-level network knowledge more accessible, allowing teams to move faster while maintaining confidence in the results.
The result is less time translating questions into queries and instant access to data required to solve network and security issues.
Forward AI is designed to streamline and accelerate common tasks for NetOps and SecOps teams, leading to significant business outcomes.
Not Another Chatbot
Forward AI isn't a documentation search tool or a generic chatbot connected to your environment.
Forward AI is production-grade because it operates on the most accurate and comprehensive network model available. It unifies all four tiers of network data hierarchy, which is essential for reliable agentic execution:
- Raw Data (configs, state, performance statistics)
- Normalized Data (vendor-agnostic model)
- Behavioral Data (actual forwarding behavior)
- Contextual Data (intent, dependencies, end-to-end semantics)
This full-stack approach ensures that the agent's actions are verifiable and trustworthy, unlike demo-grade solutions that lack accurate modeling and reliable verification.
It's built directly into Forward Enterprise and has access to the same data and applications engineers already use every day:
- Path Search
- Network Query Engine (NQE)
- Device inventory
- Vulnerability data
- VLANs, VRFs, interfaces, and subnets
- Snapshot-specific network state
Ask a question in natural language. Forward AI gathers the relevant information and returns an answer grounded in your network data.
Anthropic Claude Sonnet 4.5 and Haiku 4.5 provide the reasoning layer, while Forward's network digital twin provides the validation.
What Makes Forward AI Different?
Many AI tools generate answers.
Forward AI generates answers grounded in Forward Enterprise's network digital twin—a mathematically accurate model of your production network.
That's a critical distinction.
The AI determines how to answer the question. Forward Enterprise determines what the answer actually is and ensures the accuracy and trust of the results it returns.
Every answer is tied back to network data and analysis that can be inspected and validated, helping teams understand not just what the answer is, but why.
By combining AI reasoning with trusted network analysis, Forward AI delivers natural language interaction without sacrificing accuracy or explainability.
Forward AI combines AI reasoning with the mathematical certainty of a network digital twin. AI results are deterministic, not probabilistic with Forward’s digital twin.
How It Works in Practice
Getting started takes less than a minute:
- Open Forward AI on the left menu of the platform
- Start a new conversation.
- Ask your question.
Every conversation is tied to a specific snapshot, ensuring answers are based on a known point-in-time network state.
Forward AI also maintains context throughout the session, so investigations can evolve naturally.
For example:
Which devices are affected by this CVE?
Followed by:
Which of those are internet-facing?
Then:
Generate an NQE query so I can validate the results.
Instead of rebuilding searches, you simply continue the conversation.
What You Can Do with Forward AI Today
1. Troubleshoot a ServiceNow incident
Reference an existing ServiceNow ticket and
- Automatically troubleshoot end-to-end path connectivity or security posture
- Identify the cause of the problem
- Update the ticket with the contextual data discovered by Forward digital twin
2. Analyze Network Paths
Ask questions such as:
- Is there a path from endpoint A to subnet B?
- Why is traffic taking this route?
- Which devices are on the forwarding path?
3. Generate NQE Queries
Describe what you're looking for and let Forward AI generate the query.
For example:
Generate an NQE query that shows all interfaces with mismatched MTUs.
4. Understand Network Behavior and Configuration
Quickly answer questions like:
- What VLANs exist on this device?
- Which VRFs are configured?
- Show me all devices running IOS-XE.
- How is traffic expected to traverse this environment?
5. Investigate Vulnerabilities
Identify affected infrastructure by asking:
- Which devices are impacted by this CVE?
- Which systems are running vulnerable software?
6. Continue the Investigation
The real power comes from multi-turn conversations.
Each answer becomes the starting point for the next question, helping engineers move naturally through troubleshooting, validation, and analysis workflows.
Built for Security, Privacy, and Trust
Forward AI is read-only.
It cannot make changes to your network—a deliberate design choice that keeps operational decisions in the hands of engineers.
Customer prompts and responses are:
- Never retained
- Never used to train or fine-tune models
- Never shared outside your session
For network and security teams operating in regulated environments, privacy, transparency, and control are foundational requirements—not afterthoughts.
Looking Ahead
Forward AI is available today for SaaS customers running Forward Enterprise 26.5 with the AI feature enabled at the organization level.
Access to specific capabilities follows your existing license entitlements, including tools such as NQE and Path Search.
This first release focuses on helping engineers explore, analyze, and understand their networks through natural language. We're already working on expanding those capabilities with:
- On-premises support
- Bring Your Own Model (BYOM)
- Historical change analysis
- Blast radius analysis
- Exposure analysis
- Intent troubleshooting and diagnostics
- Expanded AI-assisted workflows
Forward AI is also laying the foundation for broader AI-assisted and agentic workflows, enabling trusted network intelligence to become part of larger operational processes.
Just as important, Forward AI is designed to be transparent about its capabilities. If you ask it to perform a task it doesn't currently support, it will tell you directly rather than attempting to generate an answer.
This is our first release—not our final vision.
We believe the future of network and security operations is conversational, explainable, and grounded in trusted data.
The network already contains the answers. Now you can simply ask for them.
Try Forward AI Today
If Forward AI is enabled for your organization, start a conversation, explore your network, and see how quickly you can move from question to answer.
We're excited to see how customers use Forward AI in real-world environments—and we're just getting started.
Have questions, feedback, or early use cases to share?
Drop them in the comments below. We'd love to hear what you're discovering and what you'd like us to build next.
